A rant : Magic (possibly Spirituality to)

[Kai]

He does this because there has been enough anecdotal evidence about 'belief' influencing survival rates,

Rubbish.  Anecdotes are not evidence.

No.  But they can indicate patterns.  Exploration of patterns can lead to evidence of causal factors.  Isn't recognition of patterns that lead to evidence that indicates causal factors pretty much the basis of scientific inquiry? (Newton, apple, etc?)

They can indicate patterns, but how well? All I have to work on is someones word, no physical measurements.

Exploration of patterns shows evidence of patterns, period. It takes INFERENCE (hypothesis) to go beyond that. A cladogram doesn't indicate evolutionary relationships, it only indicates a pattern of homologies, which are all individually hypotheses of sameness. From that, I generate a hypothesis of the reality, but that may not be the reality at all. The point is, you can never really KNOW the causal factors, which means you want the best fucking pattern evidence you can get to derive your inferences. Anecdotes don't cut it.

[The Good Reverend Roger]

Anecdotes are crap. If anecdotes were still considered as good as physical evidence we'd still be back in the middle ages.

I don't see any evidence in this thread that anyone suggested that anecdotal evidence was the equivalent of physical evidence.

He does this because there has been enough anecdotal evidence about 'belief' influencing survival rates, both positively and negatively,  to use the tool even though he has almost no way to explain it's potential, sometimes, not always replicable, efficacy. 

[Bebek Sincap Ratatosk]

He does this because there has been enough anecdotal evidence about 'belief' influencing survival rates,

Rubbish.  Anecdotes are not evidence.

No.  But they can indicate patterns.  Exploration of patterns can lead to evidence of causal factors.  Isn't recognition of patterns that lead to evidence that indicates causal factors pretty much the basis of scientific inquiry? (Newton, apple, etc?)

Pattern "recognition" can mislead you to a law of fives situation too.

Sometimes it can, sometimes it can indicate actual useful information...

I don't think Singer is arguing that all pattern recognition = Sciency Troof.

He does this because there has been enough anecdotal evidence about 'belief' influencing survival rates,

Rubbish.  Anecdotes are not evidence.

Of course they are... they may not be evidence the kind of evidence accepted within scientific models or court cases... but thats only because those systems/models have specific definitions of evidence as it applies to their models.

There was evidence that rocks fell out of the sky in the 1700's. It was all anecdotal and therefore the royal Society called it rubbish and stated that there were no rocks that fell from the sky.

The anecdotes were evidence, but they ignored it.

Anecdotes are crap. If anecdotes were still considered as good as physical evidence we'd still be back in the middle ages.

Anecdotes are not crap. They are certainly NOT objectively observed data, but that doesn't make them crap. WTF is with this IS/IS NOT thinking? Anecdotes are tremendously useful as long as they're seen as anecdotes and not objective observations. Both have value, the latter has MORE value, but the latter isn't always possible or available.

The royal society was misguided. They should have said, "There is currently no physical evidence to suggest rocks falling from the sky, but were we to be presented with  good evidence, such a claim would be acceptable." This was all before the philosophy of science really got going, so I'd go easy on them for that.

Actually they should have said "We can't substantiate your claims, but we'll send Mr. Hooke, Mr. Wren and Mr. Wilkins out in the evenings to watch for this since so many individuals have reported similar experiences."

Anecdotes = evidence that MAYBE there's something worth investigating. Objective observation = evidence that provides some detail about what is being investigated.

Simply labeling anecdotes as useless leaves us missing an incredibly useful (though often misused) tool. Simply, scientific observation cannot be everywhere, all the time. Thus we depend on individual observation (anecdotes) as a starting point for many kinds of investigation... they are not what we should base any conclusion on, or consider to carry weight... but anecdotes are evidence and they are useful.

[The Good Reverend Roger]

Anecdotes = evidence that MAYBE

wut   

And IS/IS NOT is what science is all about.  It's provable and repeatable, or it isn't.

[singer]

Pattern "recognition" can mislead you to a law of fives situation too.

I think that's what I like so much about the law of fives.  It is one of the tools that can be used very successfully when following LMNO's advice to "cut it out".  But so much of scientific advancement is sparked from intuitive leaps that further the collective understanding... all the testing and proving comes after the formulation of a hypothesis, so it seems a little counter-productive to stagnate by endlessly replicating the provable when the next big chunk of the puzzle is more likely to be out there with all the other black swans, just beyond present comprehension.

[The Good Reverend Roger]

Blarg.  I can't take any more of this new age crap.

[The Good Reverend Roger]

I knew this all reminded me of something:

http://www.principiadiscordia.com/forum/index.php?topic=18590.0

[LMNO]

Hold on...


Using Orton Nenslo's Lobster Principle here, which TGRR brought up and Singer agreed with:



if 1000 people eat a Magical Cancer Muffin, and 1 person gets healed for no apparent reason other than the MCM...



THAT IS NOT CORRELATION.  THAT IS COINCIDENCE.


If 501 people (that is, greater than half) had spontaneous remission, THAT would be correlation.



You guys are worse than the wiccans sometimes, because you should know better.

[Bebek Sincap Ratatosk]

Anecdotes = evidence that MAYBE

wut   

And IS/IS NOT is what science is all about.  It's provable and repeatable, or it isn't.

Why are you talking about science? Scientific models as I already stated to not (by definition) model anecdotal evidence. Therefore, one should not use the scientific model to discuss the value of anecdotes.

[The Johnny]

Pattern "recognition" can mislead you to a law of fives situation too.

I think that's what I like so much about the law of fives.  It is one of the tools that can be used very successfully when following LMNO's advice to "cut it out".  But so much of scientific advancement is sparked from intuitive leaps that further the collective understanding... all the testing and proving comes after the formulation of a hypothesis, so it seems a little counter-productive to stagnate by endlessly replicating the provable when the next big chunk of the puzzle is more likely to be out there with all the other black swans, just beyond present comprehension.

Im sort of afraid to jump into this thread, theres so much ambiguous terminology floating around and millions of possibilities of misinterpretation.

[The Good Reverend Roger]

You guys are worse than the wiccans sometimes, because you should know better.

It pains me, LMNO.  It really does.  It makes me hate people even more, because even the ones that ought to know better put blindfolds on themselves and run around hollering about the advantages of blindness.

Goddammit, I fucking hate people.  Every fucking one of them.  I can't fucking stand it.  Oh, Goddammit.

[Bebek Sincap Ratatosk]

You guys are worse than the wiccans sometimes, because you should know better.

[singer]

Anecdotes are crap. If anecdotes were still considered as good as physical evidence we'd still be back in the middle ages.

I don't see any evidence in this thread that anyone suggested that anecdotal evidence was the equivalent of physical evidence.

He does this because there has been enough anecdotal evidence about 'belief' influencing survival rates, both positively and negatively,  to use the tool even though he has almost no way to explain it's potential, sometimes, not always replicable, efficacy. 

Nope.  I said he used the tool based on the anecdotal evidence in conjunction with the other tools in his healing arsenal.  I did not suggest that he said "belief is enough, let's just dump this chemo down the drain".  But, based on anecdote alone (albeit many many many anecdotes) the Dr. chooses to encourage patient 'belief' as an augmentation to the other tools at his disposal.  I would be very surprised to find a Dr. that said "I don't think this will work, but I'm gonna give it to you anyway.", and I know some oncologists that spend a whole lot of time teaching each other how to sound and look hopeful so as not to shake their patients faith in the treatment, so it would seem that a patient's faith in the treatment must have an important role to play there somewhere, otherwise they wouldn't waste their very expensive time on it.

[singer]

Hold on...


Using Orton Nenslo's Lobster Principle here, which TGRR brought up and Singer agreed with:



if 1000 people eat a Magical Cancer Muffin, and 1 person gets healed for no apparent reason other than the MCM...



THAT IS NOT CORRELATION.  THAT IS COINCIDENCE.


If 501 people (that is, greater than half) had spontaneous remission, THAT would be correlation.



You guys are worse than the wiccans sometimes, because you should know better.

Is that actually the definition of correlation?  More than half?  What is 2000 people eat a magic muffin and 2 are healed for no apparent reason?  2 coincidences?  How many individual coincidences does it take to be a correlative pattern (not being a smart-ass here, I really want to know if there is an answer.)

[The Johnny]

66.7%

2/3